Expanding the molecular toolkit for study of Cadophora gregata, causal agent of Brown Stem Rot of soybean

Abstract

Chapter 2: Time Course Experiments Probe the Etiology of Brown Stem Rot of Soybean Abstract Cadophora gregata is a poorly understood ascomycete plant pathogen that causes brown stem rot (BSR) in soybeans and a few other legume species. BSR of soybean is economically significant and prevalent in the Upper Midwest of the United States. Existing BSR research has the limitation of assessing only a limited timeframe of BSR development, potentially missing key aspects of the biology governing disease development. It is also unclear when in the developmental time frame current methods, such as traditional phenotyping and qPCR, are first able to measure BSR symptoms or pathogen DNA. Further, it is not known whether measurements of pathogen DNA with qPCR can reliably distinguish between susceptible and resistant varieties challenged with C. gregata. In order to address these gaps in knowledge, time-course experiments were conducted in the field and greenhouse, allowing investigation of BSR development, methods for quantifying BSR, and the lifecycle of C. gregata. In controlled environments, qPCR analysis revealed that C. gregata DNA generally becomes detectable within 22 days post inoculation in the lower stem of infected plants, and that the level of C. gregata DNA in soybean stems correlates to the BSR symptom severity as measured at 9-weeks post inoculation. Differences in C. gregata DNA quantity between susceptible and resistant varieties were not statistically significant when measured at a single time point, but the timeframe for detecting the pathogen varied significantly between varieties. In-planta microscopy observations of C. gregata matched previous descriptions in the literature with hyphae growing at variable density in the xylem of the host, but we also observed tyloses in infected plants not seen in mock-inoculated plants. Symptom severity using traditional methods increased until the end of the experimental timeframe. Cadophora gregata was not detected until the middle of August in plants in field experiments, challenging the current model for the lifecycle of the pathogen which stipulates that it infects soybeans in the seedling stage. Finally, we successfully converted the qPCR assay to a digital droplet PCR assay with some notable advantages and limitations. Overall, this study provides valuable insights into BSR development and detection and highlights the need for further investigations into the lifecycle of C. gregata. Chapter 3: A ‘Hifi only’ approach yields high quality draft genome sequence of the A and B types of Cadophora gregata, causal agent of Brown Stem Rot of soybean Abstract Cadophora gregata is an economically important phytopathogen which causes brown stem rot (BSR) in soybeans, adzuki beans, and mung beans. Cadophora is a diverse and widespread genus of ascomycetes, which in addition to C. gregata also contain numerous symbiotic and free living fungi, many of which hail from unusual or extreme habitats, as well as endophytes and plant pathogens. Little is known about Cadophora in general. In the case of C. gregata key questions remain about the life cycle, interactions with the host, and the genetic basis for differences between isolates, which have been grouped into two types (type A and B) based on observations of host phenotype and an INDEL mutation in the intergenic spacer region of the ribosomal RNA gene cluster. In recent years four Cadophora genomes have been produced, but not of C. gregata. However, all of these assemblies were performed using short read sequencing only, and evidence-based gene annotation was based on only one total RNA sample. In this study we present the first full genome sequences of type A and type B isolates of C. gregata. We leveraged accurate long reads from PacBio Hifi to assemble these genomes in one step, without the inclusion of illumina short reads. For evidence-based gene annotation, we included 16 separate samples from axenic cultures and infected soybean plants. The two genomes, generated in parallel, were similar. The A type isolate MNR1 had an assembly with 95 contigs. The B type isolate MN4 had 42 contigs. The contig N50 was 2.5 Mbp for both isolates. Both assemblies were similar in length (47 Mbp) and had a similar total number of gene models represented (15k). The BUSCO score for both genomes was above 99%. Differences were identified between the A and B type genomes, such as presence absence variations in various transcription factors, in gene count for certain gene families, and differences in the total number of tRNA genes found. The overall consistency and quality of PacBio hifi data in generating high quality genome assemblies for these related taxa makes this dataset useful for future comparative genomics work.